import html2text
import jsonlines
class Dataset(object):
def __init__(self, gt_file, img_name, results):
self.gt_file = gt_file
self.img_name = img_name
self.results = results
self.pre_list = []
self.gt_list = []
def __len__(self):
return [len(self.pre_list), len(self.gt_list)]
def get_pre_list(self):
pre_xml = self.results['tableList'][0]
self.pre_list = parse_pre_str(pre_xml)
return self.pre_list
def get_pre_structure(self):
pre_xml = self.results['tableList'][0]
# print('gt', pre_xml)
pre_html = html2text.html2text(pre_xml) # str
return pre_html
def get_gt_list(self):
with jsonlines.open(self.gt_file, 'r') as rfd:
for data in rfd:
gt_xml = data['gt']
# print(gt_xml)
self.gt_list = parse_gt_str(gt_xml)
return self.gt_list
def get_gt_structure(self):
with jsonlines.open(self.gt_file, 'r') as rfd:
for data in rfd:
gt_html = html2text.html2text(data['gt']) # str
return gt_html
gt_html = 'Error:并未找到需要该图片的标注信息!'
return gt_html
def parse_gt_str(text):
text = text.replace('| ', '')
text = text.replace(' | ', '')
text = text.replace(' | ', '')
text = text.replace('', '')
text = text.replace('', '')
text = text.replace('', '')
text = text.replace('', '')
text = text.replace('', '')
text = text.replace(' ', '')
text = text.replace(' | ', '')
text = text.replace('', '')
# print('gt', text)
text = text.replace('', '')
text = text.replace(' | ', '*')
text = text.replace('', '')
return text.strip('
').split('')
def parse_pre_str(text):
text = text.replace('', '')
text = text.replace(' | ', '')
text = text.replace(' | ', '')
text = text.replace('', '')
text = text.replace('', '')
text = text.replace('', '')
text = text.replace('', '')
text = text.replace('', '')
text = text.replace(' ', '')
text = text.replace(' | ', '')
text = text.replace('', '')
# print('pre', text)
text = text.replace('', '')
text = text.replace(' | ', '*')
text = text.replace('', '')
# return text.strip('
').split('')
return text.strip('').split('')